“Information is the oil of the 21st century, and analytics is the combustion engine.”

– Peter Sondergaard, SVP at Gartner Research

Data is awesome. More specifically, what some companies are doing with data today is awesome. We’re seeing advances ranging from powerful voice recognition for instant concierge services to autonomous vehicles and AI-based pharmaceutical testing. Advanced machine learning is optimizing transportation logistics, energy management, agriculture yields, and thousands of other large and small applications. Leaders are finding innovative ways to analyze and use data to make their products and services better.

Becoming a data-driven company isn’t always easy, and many companies struggle to advance their data intelligence capabilities to keep pace with the opportunities they foresee. Business and technology leaders are dealing with a number of disruptions that make it difficult to successfully manage and harness new data-driven capabilities efficiently across the enterprise. Some of the challenges they face include:

  • The vendor landscape continues to become more complex and fragmented – Just take a look at one recent big data vendor landscape.
  • Corporate data continues to fragment – Hosted SAAS solutions are replacing key in-house applications, and data that used to live within the boundaries and control of the company’s data center now reside on someone else’s servers. In the networked economy, transactions become more complex, and any given interaction can undergo numerous callouts to external APIs and handoffs to third-party systems. The tools to collect, manipulate, and aggregate data continue to improve, but each new touchpoint requires an effort to setup and fold it into your data intelligence platform. And this assumes that your vendor agreements are even set up correctly to ensure you can access the data you desire with the frequency you want it.
  • The consumption pattern is changing, too – Dashboards, predictive output, and real-time analysis are now being embedded as functionality into solutions, not just presented in standalone reporting systems. Product teams and Operations groups want to pull real-time results into composite applications, and delivering to highly-interactive systems (like virtual reality) requires unique infrastructure and processes.
  • Non-corporate data is a different beast – The “Internet of Things” introduces vast new layers of “non-corporate” data acquired from sensors embedded in everyday items, such as information about its users’ activities, locations, and habits not directly related to any explicit business transaction. Social media activity, browsing behavior, and other “observable” data is being siphoned, correlated, and analyzed with only indirect consent. Public expectations about the handling, stewardship, and internal use of this data continues to fluctuate, challenging companies to adapt in order to remain in good standing with their customers.
  • Decentralized and autonomous decisions need new verification processes – New predictive tools and machine learning algorithms can directly drive system behavior and render customer-facing business decisions. Companies need to set up processes to periodically review the decisions generated and confirm they are still aligned with desired behavior, especially in self-learning systems.
  • Responsibility for delivering data-driven capabilities is no longer just part of IT – Expectations continue to change about the role of CMOs to drive data-based innovation, and HBR reports nearly 10% of marketing budgets at larger organizations are being spent on marketing analytics tools. Other business groups are similarly prototyping and deploying tactical solutions for their function, which often become a de-facto part of the production ecosystem. This can generate tension between IT and business groups about technology decisions and ongoing system management (not to mention the related challenges in how technology budgets get allocated).

None of these are insurmountable, and there are many examples of companies adapting well to these types of changes. Successful organizations see these systemic disruptions and know they need a strategy that goes beyond just identifying new technology to implement in isolated pockets of the company. A successful data-capability strategy also needs to address elements, such as:

  • Organizational structure and skillsets needed to tackle the next level of complex analysis and how this expertise is leveraged across the enterprise
  • Process touchpoints and transitions, such as what data governance needs to be in place (there are many facets to just this question!) or how analytics will fuel product planning
  • Alignment of data-based capabilities into the broader corporate strategy

Most importantly, corporate leaders need to discuss the data-driven capabilities and needs of their company cohesively and deliberately. The current disruptions are not specific to just a particular function within the company. Nor is a single figurehead likely to develop a solution that meets the varied needs of the broader enterprise. Hence, organizations need a common data intelligence strategy. By understanding what makes data intelligence hard and developing a cohesive, long-term plan that addresses these challenges, the management team is more likely to end up building a data analytics hot rod instead of just a go-cart.